Quickly design and prototype API's.
Describe your data structure using the Hash format.
Get JSON test data automatically based on your input.
A classical example would be something like this - you're a mobile engineer and you need a new API but it takes forever for the backend guys to build it so you are pretty much blocked until they have something to show.
Using LightAPI you can prototype the API in collaboration and work independently while your team mate can code the real API using the prototype you guys designed as a baseline and you the iOS engineer can also start building and simply replace the endpoint when the official API is ready!
http://localhost:8080/trip/users.json?response={id:number,url:string,name:string,email:email,%20date:date}&rows=3
[
{
id: 4522344254,
url: "http://koch.biz/emelie.raynor",
name: "officiis",
email: "iva.muller@cronin.info",
date: "2017-11-16"
},
{
id: 1136377739,
url: "http://bahringer.co/felipa_dooley",
name: "voluptatem",
email: "easton.dietrich@trantow.biz",
date: "2017-11-11"
},
{
id: 8814665917,
url: "http://kerlukekeebler.com/enoch.williamson",
name: "unde",
email: "rosemary_wyman@dubuquekuhic.com",
date: "2017-11-08"
}
]
git@github.com:eduardosasso/lightapi.git
docker-compose up --build
http://localhost:8080
docker-compose logs -f
remove all images
docker rmi -f $(docker images -q)
remove all containers
docker rm -f $(docker ps -a -q)
get shell access
docker-compose exec app sh
Create a droplet on DigitalOcean
docker-machine create lightapi --driver=digitalocean --digitalocean-access-token=<token>
Deploy to production
eval $(docker-machine env lightapi)
docker-compose -f docker-compose.yml -f docker-compose.prod.yml up --build
eval $(docker-machine env -u)